Are Bottlenecks Endemic to the Grid?
Blackouts, like the one in Southern California in September, are a fact of life
Hi, this is Steven Cherry for IEEE Spectrum’s “Techwise Conversations.” This is show number 66.
On Thursday, September 8th, 1.4 million households in Southern California lost power in a outage that started with some routine maintenance work on a power line in Arizona. Another 56 000 customers back in Arizona, and a few hundred in the Mexican state of Baja California, were also affected. Power was restored over the next two days.
At the time, the headlines were dominated by the idea that a single electrical worker could be to blame. That misses the main story almost entirely. Why isn’t the grid robust enough to recover from a simple mishap on a single high-voltage line connected to a single substation?
My guest today is here to help answer that question. Roberta Gamble is the director for energy and environment at Frost & Sullivan, an analytical research firm that specializes in emerging technologies and markets. She joins us by phone from Mountain View, California. Roberta, welcome to the podcast.
Roberta Gamble: Thank you, Steven.
Steven Cherry: Let’s start with the outage itself. It began in Yuma, Arizona, a city of about 100 000 that lies right on the border with California. The substation there did shut down, but it didn’t stop there. What happened?
Roberta Gamble: Well, it should have shut down. It certainly should have stopped right there, and it’s certainly puzzling. I think, from what I understand, we’re still trying to figure out what happened exactly—why this happened there or frankly anywhere else, because we do have smart grid capabilities on our transmission lines in this country, by and large, like 80 to 90 percent transmission, because it is such a critical part of our infrastructure, and so many communities ride on that one transmission corridor, for example. So, yeah, it’s a great question. I don’t think we’ll know for a little bit what exactly did happen, what possibly failed, or what precautions hadn’t been adequately put in place to keep an event like this from happening.
Steven Cherry: Computer network architects talk about distributed networks. The Internet is a distributed network; it’s designed to resist failure by having lots of alternative paths between any two points. I gather that’s a lot easier to do with packets of data than with voltages and electrons.
Roberta Gamble: Well, it is and it isn’t. I mean, you can have distributive generation, which has been a growing market for many years now, where you’re generating electricity at the point at which you’re using it. In other words, you have a backup generator at your house, at your building, or even a prime power, so something that’s providing the most electricity on a continuous basis. You may have a small gas turbine, even at a very large industrial facility, that’s providing electricity where it’s being used. Solar panels are a form of distributed power, here again generating electricity, when you’re generating that electricity where it’s being used and you don’t need that distribution system. The issue is twofold: One, you’re rarely going to be able to be completely off the grid. It’s possible—you see homes, like vacation homes in Maine, for example, that may be relatively off the grid; they just have a power backup generator—a power generator, I should say—in their home. And the second issue is, since you’re rarely off the grid, you’re almost always going to have to have at least supplemental, if not your main power, coming from the grid—there’s safety concerns. So it would be to go to a distributed solution from the centralized hub-and-spoke solution that we have now, where a central power delivers electricity out to a greater area; it would be a complete revamping of our energy system. I mean, it would probably not be something that we would be likely to see—ever. There’s a lot of sunken costs in large power plants, and it’s a relatively cleaner solution than, let’s say, running a diesel generator for the most part, if you’re looking at certain types of power plants running, say, multiple numbers of diesel generators around a distributed area. So in terms of—I mean, it’s possible from a physical standpoint; there’s nothing that says that you couldn’t do it, but costs and just the fact that we have this existing centralized power distribution system makes it unlikely that we would ever move to a predominantly distributed solution in this country.
Steven Cherry: So I guess there’s no simple answer to the question of whether wind power and solar power make the grid more robust or just the opposite?
Roberta Gamble: Yeah, that’s an interesting question. You know, it’s intermittent power, so either you have power that’s fluctuating for whenever the sun shines or the wind blows on the grid, or more likely, you have some sort of storage at the wind or solar generation points so that you’re not on and off all the time on the grids and you’re, you could almost say downloading power—a certain amount of power during the day, but you’d have to have a lot of energy storage and a lot of wind or solar to be able to have that consistent. You’d have to have a lot more installed megawatts of wind power and then quite a bit more of storage to be able to have a consistent flow. So, yeah, there’s some electricity that’s put on the grid through wind and solar that’s not on a consistent basis, but on the other hand, our usage isn’t consistent either. So you try to time those—that generation well; you can’t really time the generation, but the releasing of that electricity for peak times when we’re using the most energy. There are other ways to store energy and if utilities, and homeowners, and buildings, and commercial entities were to store more of their power at night and release more of it during the day, that could help with some of the load as well because again this is—as a country, we generate a lot of electricity, we generate more than enough electricity. The problem is, we are not always able to transmit the electricity as efficiently as we’d like, and we don’t always generate enough during the peak times when we need it. But if you look at an absolute amount, we have enough kilowatt hours generated to be able to support all our needs; it’s just the issue is the timing as well. So I wouldn’t say solar or wind add so much more complication to the grid. I think it’s just more the demand profiles and being able to manage any source of electricity that you have going across the grid.
Steven Cherry: And if electric cars become a significant portion of our transportation, that has the potential to, in effect, make the grid a little less distributed, really, because right now we’re distributing a certain amount of transportation energy across 200 million vehicles, and we’re potentially putting that onto the grid.
Roberta Gamble: Yeah, it could be less distributed; it could even be more distributed, because electric vehicles are a source of power storage as well as power usage—you know, you have a big battery on there is what you’re using. It could work to destabilize the grid a bit more if we have a really large number of electric vehicles come online and we’re not well prepared for that. There are ways you could feasibly manage the charging and the discharging of electric vehicles that you could help balance power usage. What you don’t want is you don’t want 2 million electric vehicles in a given area charging at 3 o’clock in the afternoon in the middle of the summer, when your air conditioning costs are at their highest. But maybe there’s a way to charge the vehicles at night, when the power is lowest cost and is most abundant. And then, when you plug it in—so you charge your vehicle at night, you go to work, you plug it in; the electric utility needs electricity, let’s say between 12 and 3 o’clock in the afternoon. They actually discharge electricity from your vehicle, so they’re using it as an energy storage medium. The only catch to that is they better charge you up before you leave the office at 6 o’clock, and there’s also questions of whether or not you want to give the utility that much control. What if there’s an emergency? What if there’s a reason that you need your vehicle? Do you want your utility to go in and use your vehicle as an energy storage medium?
Steven Cherry: So it sounds like the grid is just going to get more and more complicated.
Roberta Gamble: Absolutely.
Steven Cherry: Does that mean that blackouts and cascading failures are just sort of inevitable, that we just sort of expect them every few years?
Roberta Gamble: Well, I’ve seen research that states that we have blackouts more often. I wouldn’t say they’re going to become more inevitable, because with these other complicated factors in the grid, and let’s not forget that we also have the potential for third parties and energy management services, so you could have—you’ve got this backup power; you’ve got solar panels on roofs that could be putting electricity back on the grid; you have electric vehicles, but then you also have, say, a third-party provider come in and manage and monitor your smart-home electricity usage, and maybe they’re going to want some of the utility information as well, and so there’s other parties that could be coming into play. Now with that comes the need for the grid to become increasingly more intelligent, and we’ve seen a lot of movement towards utilities smartening up their grid. The American Recovery and Reinvestment Act that provided almost 5 billion dollars to utilities and matching funds to—it was about 4.8, I believe—to add automated meter infrastructure AMI and smart meters. The utilities jumped on that; the funds were requested almost immediately, and a lot of the utilities in fact missed out on that funding. So I think it’s not inevitable. Is it likely? Possibly? Will utilities be able to do enough in time to mitigate more of these problems? Maybe not. So I wouldn’t say it’s inevitable, but it’s a possibility.
Steven Cherry: Very good. Well, maybe that’s a good place to leave it. Thank you very much.
Roberta Gamble: Thank you.
Steven Cherry: We’ve been speaking with Roberta Gamble, director for energy and environment systems at the analytical research firm Frost & Sullivan, about the problem of cascading failures in the electrical grid and what can be done about them. For IEEE Spectrum’s “Techwise Conversations,” I’m Steven Cherry.
This interview was recorded 12 September 2011.
Audio engineer: Francesco Ferorelli
Follow up on Twitter @Spectrumpodcast
NOTE: Transcripts are created for the convenience of our readers and listeners and may not perfectly match their associated interviews and narratives. The authoritative record of IEEE Spectrum’s audio programming is the audio version.